Association of chronic pain and depression with brain structure

Pre-processing and QC - brain structure data:

• Outliers removed (>5 SD)

• Global DTI measures calculated using PCA

—————————————————————————————————————————-

Associations between DNAm scores and brain structure data

Models:

- GLM for unilatearal brain phenotypes - LME for bilateral brain phenotypes where measures of hemisphere were treated as repeated measures - No hemishphere interactions on association between DNAm score and bilateral phenotype observed

Covariates: - Sex - Age - Age^2 - BMI - UKB imaging assessment centre - ICV (only for measures of brain volume)

Multiple comparision correction: FDR - n = number of brain structure phenotypes per imaging subcategory

Sample demographics

Main Results - Brain structures

Effect size graph - Comorbidity in full sample

P-value graph - Depression - full sample

## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.

P-value graph - Depression - sex-stratified sample

P-value graph - Chronic Pain Status - full sample

P-value graph - Chronic Pain - sex-stratified sample

Cortex map: chronic pain

Subcortex volume map: chronic pain

## merging atlas and data by 'region'
## merging atlas and data by 'region'
## merging atlas and data by 'region'
## merging atlas and data by 'region'

Cortex map: depression

Subcortex volume map: depression

## merging atlas and data by 'region'
## merging atlas and data by 'region'
## merging atlas and data by 'region'
## merging atlas and data by 'region'

DTI: MD map

DTI: FA map

Supplementary Results

Effect size graph - Chronic pain status in full sample

Effect size graph - Chronic pain status in sex-stratified samples

Effect size graph - Probable Recurrent Major Depressive Disorder in full smaple

Effect size graph - Probable Recurrent Major Depressive Disorder in sex-stratified sample

Effect size graph - Balanced datasets

Number of significant hits

sum(res_comorbidity_brain_structure_all_save$p.adjust < 0.05)

## [1] 94

sum(res_comorbidity_brain_structure_female_save$p.adjust < 0.05)

## [1] 88

sum(res_comorbidity_brain_structure_male_save$p.adjust < 0.05)

## [1] 4

sum(brain_structure_CP_all$`Adjusted p-value < 0.05` == "Yes")

## [1] 50

sum(brain_structure_CP_female$`Adjusted p-value < 0.05` == "Yes")

## [1] 38

sum(brain_structure_CP_male$`Adjusted p-value < 0.05` == "Yes")

## [1] 14

sum(brain_structure_MDD_all$`Adjusted p-value < 0.05` == "Yes")

## [1] 26

sum(brain_structure_MDD_female$`Adjusted p-value < 0.05` == "Yes")

## [1] 29

sum(brain_structure_MDD_male$`Adjusted p-value < 0.05` == "Yes")

## [1] 2

Number of significant hits - balanced vs unbalanced

cat("Unbalanced comorbidity:")

## Unbalanced comorbidity:

sum(res_comorbidity_brain_structure_female_save$p.adjust < 0.05)

## [1] 88

sum(res_comorbidity_brain_structure_male_save$p.adjust < 0.05)

## [1] 4

cat("Balanced comorbidity:")

## Balanced comorbidity:

sum(res_comorbidity_brain_structure_balanced_f$p.adjust < 0.05)

## [1] 22

sum(res_comorbidity_brain_structure_balanced_m$p.adjust < 0.05)

## [1] 2

cat("Unbalanced CP:")

## Unbalanced CP:

sum(brain_structure_CP_female$`Adjusted p-value < 0.05` == "Yes")

## [1] 38

sum(brain_structure_CP_male$`Adjusted p-value < 0.05` == "Yes")

## [1] 14

cat("Balanced CP:")

## Balanced CP:

sum(res_CP_MDD_brain_structure_balanced_CP_f$p.adjust < 0.05)

## [1] 61

sum(res_CP_MDD_brain_structure_balanced_CP_m$p.adjust < 0.05)

## [1] 27

cat("Unbalanced MDD:")

## Unbalanced MDD:

sum(brain_structure_MDD_female$`Adjusted p-value < 0.05` == "Yes")

## [1] 29

sum(brain_structure_MDD_male$`Adjusted p-value < 0.05` == "Yes")

## [1] 2

cat("Balanced MDD:")

## Balanced MDD:

sum(res_CP_MDD_brain_structure_balanced_MDD_f$p.adjust < 0.05)

## [1] 43

sum(res_CP_MDD_brain_structure_balanced_MDD_m$p.adjust < 0.05)

## [1] 3